Want to know if this job is worth applying to?
São Paulo, Brazil, Brazil
Onsite
Overall, Purpose of the Job
We are hiring for the AI & Technology Solutions (ATS), Clinical AI & Technology Innovation (CAITI) – Technical Support Engineering (TSE) team. This team is responsible for operating, supporting, and continuously improving Agentic AI products running in regulated production environments.
The role focuses on first‑line and second‑line technical support for multi‑agent AI workflows, ensuring that AI agents, orchestration layers, data pipelines, and downstream integrations operate reliably, safely, and in line with enterprise and regulatory expectations.
The incumbent will act as a key interface between end users, AI product teams, and engineering, enabling rapid issue resolution, operational insights, and continuous learning across Agentic AI systems developed by the ADSC team.
Principal Accountabilities
1. Agentic AI Production Support & Operations
· Serve as the single point of contact (L1/L2) for users of Agentic AI products across multiple ADSC offerings, supporting multi‑agent workflows in live production environments.
· Monitor, triage, and resolve incidents related to:
o Agent orchestration and failures
o Tool/ skill execution errors
o Data access and pipeline issues
o Model degradation or unexpected agent behavior
· Ensure stable, predictable, and auditable operation of AI agents in regulated clinical and analytics contexts.
2. Agent Behavior Analysis & Root Cause Investigation
· Analyze agent execution logs, prompts, tool calls, and system telemetry to:
o Identify failure patterns
o Perform root cause analysis (RCA) across agent, application, and infrastructure layers
o Distinguish between model issues, orchestration logic issues, data issues, and user‑input driven failures
· Work closely with AI engineers and architects to feed operational insights back into:
o Prompt improvements
o Agent policy and guardrail refinement
o Workflow and orchestration enhancements
3. Human‑in‑the‑Loop & User Enablement
· Act as a human‑in‑the‑loop escalation layer, ensuring that:
o Agent outputs are understandable and explainable to users
o Users receive clear guidance when agent confidence thresholds or guardrails are triggered.
o Provide hands‑on guidance and troubleshooting to help users make effective use of Agentic AI products, including interpretation of agent decisions and outputs.
4. Knowledge Base & Operational Readiness
· Develop and maintain a living knowledge base covering:
o Agent behavior FAQs
o Common failure modes and resolutions
o Known limitations, guardrails, and safe‑usage patterns
· Keep documentation continuously updated to reflect:
o New agent capabilities
o Model upgrades
o Workflow changes and feature releases
5. Continuous Improvement & AI Health Monitoring
· Track and publish operational KPIs and AI health indicators, including:
o Agent success/failure rates
o Latency and throughput
o User‑reported issues vs. System‑detected issues
· Proactively identify opportunities to:
o Reduce operational toil
o Improve agent reliability
o Increase user trust and adoption
6. Cross‑Team Collaboration
· Collaborate closely with:
o Product Managers and Product Owners
o Engineering teams
o Data platform and upstream/ downstream application teams
· Participate in regular product and engineering syncs to ensure operational feedback loops are embedded into the Agentic AI Software Development Lifecycle (SDLC).
Essential Experience
· Experience across analytics, technology, or consulting, with exposure to production of AI or complex data systems.
· Hands‑on experience supporting complex, distributed production systems, preferably involving:
o AI/ML components
o Workflow orchestration
o Data‑intensive applications
· Experience working with global, cross‑functional teams in fast‑paced, service‑oriented environments
Technical & Professional Requirements
· Bachelor’s degree in Computer Science, Engineering, Information Technology, or related field
· 4–7 years of experience supporting complex production systems.
· Strong proficiency in:
o Python (for debugging, scripting, and analysis)
o SQL (for data investigation and validation)
· Familiarity or strong interest in:
o Agentic AI concepts (agents, tools, orchestration, guardrails)
o Logs, telemetry, and observability in AI systems
o Agentic libraries such as LangChain, LangGraph, LangSmith
· Exposure to big‑data or distributed technologies (e.g., Hive, Kafka, HDFS, Impala) is a plus
· Basic understanding or willingness to learn Machine Learning fundamentals, particularly as they relate to production behavior and limitations
Key Behavioral Attributes
· Strong analytical mindset with the ability to reason about AI system behavior end‑to‑end
· Clear and confident communicator, able to explain complex agent behaviors to non‑technical users
· Proactive, self‑motivated, and continuously curious about emerging AI technologies
· Comfortable operating at the intersection of users, AI systems, and engineering teams
IQVIA is a leading global provider of clinical research services, commercial insights and healthcare intelligence to the life sciences and healthcare industries. We create intelligent connections to accelerate the development and commercialization of innovative medical treatments to help improve patient outcomes and population health worldwide. Learn more at https://jobs.iqvia.com
IQVIA is committed to integrity in our hiring process and maintains a zero tolerance policy for candidate fraud. All information and credentials submitted in your application must be truthful and complete. Any false statements, misrepresentations, or material omissions during the recruitment process will result in immediate disqualification of your application, or termination of employment if discovered later, in accordance with applicable law. We appreciate your honesty and professionalism.